This is an appeal from the decree of the Circuit Court for the Eastern District of Pennsylvania, dismissing the bill of the complainant appellant, charging infringement against the defendant appellee of a certain patent, No. 743,132, granted November 3. 1903, to the Dodge Needle Company as assignee of one Arthur Currier, for an improvement in knitting machine needles. The following extracts from the specifications of the letters patent set forth the prior art as recognized by the applicant, and the character and structure of the needle devised by him for the purpose of overcoming the defects of the needles in the prior art, as described and differentiated therefrom :

“Needles for knitting-machines are commonly made by providing the body of the needle with a slot in which the end of the latch is pivotee!, a pivot pin being passed through the body of the needle and the latch for this purpose and being headed over at its ends to retain it in place. One objection to this construction is that it is very difficult to spread or head over the end of the pivot pin, so as not to leave a slight bur or roughness which will catch on the fiber of the yarn and injure the same. Various expedients have been resorted to to cure this objection, one of tbem being to insert the pivot pin into a drilled hole in the body of the needle, said pivot pin passing through the usual eye in the end of tlie latch and being shorter than the width of the body of the needle, whereby the ends of pivot pin are below the surrounding surface of the side walls of the needle. To hold the pivot pin in place, the body of the needle was swaged or up set, so that portions thereof were displaced to partially close the drilled hole therein, the displaced metal overlapping the ends *716of the pivot pin. and operating to retain it in place. This manner of securing the pivot pin in place, however, has not been entirely successful, because it is extremely difficult to locate a pivot pin centrally in the needle body, so that its opposite ends are supported uniformly in the body of the needle, as is necessary to make a perfect needle. Moreover, difficulty has been experienced in making the countersinks in the opposite sides of the needle of uniform depth, this difficulty arising from the fact that the walls of the needle are very thin and the pivot pin is apt to have its ends extended unequally into the body of the needle. It has also been proposed to secure the pivot pin in place by first making a comparatively deep countersink in the side walls of the needle and heading over the ends of the pivot pin into said countersink in such a way that the edges of the heads on the pivot pin are below the surrounding surface of the side walls of the needle. This form of needle, however, is objectionable, because the making of the comparatively large countersinks in the side walls thereof materially weakens the needle at the very point where the most strain comes thereon and where the needle needs to be the strongest. Moreover, experience has demonstrated that even though when the needle is first made according to this method, the head of the pivot pin is carried below the side walls of the needle; yet after the needle has been in use some time the pivot invariably works loose and the thin edges of the heads of the pin will project above the surrounding surface of the side walls of the needle, thus forming a bur or roughness which will .catch the fiber of the yarn. Still another way in which needles have been constructed is to provide the side walls of the needles with the usual countersinks into which the ends of the pivot pins are headed or spread, as usual. The headed ends of the pivot ifins and a portion of the surrounding side walls of the needles have then been dressed off by a revolving tool to form countersinks, so that the ends of the pivot pins are carried below the surface of the side walls of the needle. It has been found in practice, however, that a revolving tool will leave a rough edge surrounding the countersink, and such rough edge is sufficient to catch and pull the fine fibers of delicate yarns, thereby unfitting a needle constructed as above described for many classes of work. Another disadvantage in this form of needle is that in forming the countersinks by dressing off or removing the portion of the needle body the said body is thereby weakened.

“It is the object of my invention to provide a novel form of knitting machine needle which will do away with all of the disadvantages above named and which at the same time will be so constructed that there is no bur or other roughness which can engage and pull the fiber of the yarn. Accordingly I provide my needle body and latch with the usual drilled hole, in which the usual pivot pin of suitable length is inserted, and after the pivot pin has been put in place it is shortened either by cutting or grinding off or hammering down the ends of the pin until it is of a length substantially equal to the width of the needle body, the ends of the pivot pin coming substantially flush with the body. Thereafter the needle is placed between suitable dies or punches, the acting faces of which are convex in shape and are of slightly greater area than the area presented at the ends of the' pivot pin, and these punches meet the outer sides of the needle, cover the ends of the pivot pin, and shorten or swell the pivot pin in the body of the needle and in the hole In the latch thereof, the hole in the latch being enough larger than the upset pivot pin to leave the latch free to turn on the pivot pin. This operation of the punches compresses the material of the needle body to form a depression in each side of the needle and at the same time shortens or compresses the ends of the pivot pin and carries the said ends below the surface of the surrounding side walls of the needle, the ends of the pivot pin constituting and lying substantially flush with or a little below the bottoms of the depressions.

“The operation of shortening the pivot pin in the body of the needle operates to swell the same to thereby completely fill the hole in the body of the needle, whereby the pivot pin is firmly held in place and at the same time the ends of the pivot are carried below the surrounding side walls of the needle, a feature which experience has proved is necessary in this class of devices. Moreover, by forming the depressions by compression instead of by removing a part ■of the metal to form usual countersinks it will be seen that the full original strength of the needle body is preserved at the point where the said body is weakest and is subjected to the most strain.”

*717Tlie claims of the patent are as follows:

“L A knitting machine needle comprising a body having a latch mounted on a pivot pin, the said pivot pin and the side walls of the body adjacent the ends of the pin being compressed to form depressions, the compressed ends of the pins coinciding with the bottom of the said depressions.

“2. A knitting machine needle comprising a body, a latch and a pivot pin in said body upon which said latch is loosely mounted, the side walls of the body adjacent the ends of the pins being compressed to form depressions, the ends of the pins being flush with and forming a portion of the bottoms of the said depressions.

“3. A knitting machine needle comprising a body, a latch, and a pivot pin in said body upon which said latch is loosely mounted, the outer side walls of said body having that portion which surrounds and is concentric with the ends of the pivot pin compressed to form depressions, the pivot pin also being compressed longitudinally thereof and the ends of said pin when compressed being flush with and forming a portion of the bottom of the depressions.”

The following are the drawings accompanying the patent:

*718The defenses set up in the answer are: (1) Noninfringement; (2) anticipation; (3) lack of patentable novelty; (4) abandonment; and (5) public use and sale for more than two years before the application for the patent in suit was filed.

It is apparent, from reading the claims and specifications, that the patent relates to a structure characterized by its physical features, and not by the method of its construction. These physical features attach themselves to so small a part of the article or structure in question, and are, in consequence, so minute as to be incapable of recognition by the naked eye, and are measured by the hundredth, or even thousandth, part of an inch, and though the microscope and micro-photography may be resorted to for illustration, within certain limits, the imagination must still be called upon to follow the changes that are suggested by the application of known laws of matter under the conditions described. These conditions are not referred to as depreciating the value of the patent, but only by way of caution, lest we exaggerate certain features of physical structure by the lines of the merely conventional drawings of the patent, or of the exhibits in the case. It can be readily understood, from the specifications of the patent and from the testimony in the case, how difficult, as well as important, it was to so mount the latch in the slot of the needle, as that the ends of the minute pivdt pin, with their bearings in the necessarily very thin sides of the needle, should not so protrude as to catch, with their roughened or upset heads, the fiber of the yarns with which the swiftly moving needles are constantly being engaged and disengaged.

As referred to in the specifications, the prior art sought to overcome this difficulty in several ways: (1) By making the pivot pin, which passed through the eye of the latch into the sides of the slot, shorter than the width of the needle, whereby the ends of the pivot pin were below the surrounding surface of the side walls of the needle. The body of the needle was then swaged, or upset, so that portions thereof were displaced to partially close the drilled hole therein, the displaced metal overlapping the ends of the pivot pin and operating to retain it in place. (2) Another way cited to secure the pivot pin in place, was to make comparatively deep countersinks in the side walls of the needle, and to head over the ends of, the pivot pin into the countersinks, so that the edges of the heads of the pivot pin are below the surrounding surface of the side walls of the needle. (3) The other way referred to, was to provide the side walls with the usual countersinks, into which the ends of the pivot pins are headed or spread, as usual. The headed ends and a portion of the surrounding side walls were then dressed off by a revolving tool. The objection stated in the specifications to this latter method, is, that a revolving tool will leave a rough edge surrounding the countersink, sufficient to catch and pull the fine fibers of the yarn. It is also obj ected that, like the immediately preceding method, the body of the needle is weakened by removing a portion thereof to make the "countersinks.

In all these needles referred to as in the prior art, the ends of the latch pivot were secured in their bearings in the sides of the needle, by being headed or upset, whether on the surface of the sides of the *719.needle or at the bottom of a countersink formed therein. This heading or upsetting was necessarily produced by the force of tools, or of pressure applied longtitudinally to the pin, and effected a union to some extent between the end of the pin and the hole in which it was placed. The original method described in the specifications, of heading or riveting the heads of the pins on the surface of the sides of the needle, doubtless made a secure and strong emplacement of the pin, but the tendency of the bur and roughened heads on the surface of the needle, to catch the fibers of the yarn, made the difficulty to be overcome. The making of countersinks in the thin sides of the slot of the needle, by cutting away the material, by a revolving tool or other means, obviously weakened the structure at that point where strength was especially necessary. As a countersink or depression of some kind in the sides of the needle surrounding the pin holes, was absolutely necessary, in order to keep the heads of the pin away from the path of the yarn fibers, the contention of the appellant is, that it was the happy thought of the patent in suit, that, by longitudinal compression applied to the ends of the pin by opposing dies, the acting faces of which are convex in shape and of slightly greater area than that presented at the ends of the pivot pin, so as to cover an annular portion of the needle sides adjacent to and concentric with the pin heads, not only would the pin itself be shortened and thereby swollen in the holes in the sides of the needle, liras making it secure in its bearings, but, at the same time, and by the same operation, a depression on the sides of the needle would be formed by the compacting pressure of the die. A needle is thus produced with a countersink, the formation of which has not weakened the needle by cutting away any portion of the metal composing it, and the ends of the pin are tightened in their bearing holes without upsetting or heading. Oil the contrary, the heads form the bottom of the depressions, the center of the ends being slightly more.depressed than their circumferential edges. A needle thus constructed, without initial countersinks and heads spread to fill them, but only with depressions made by pressure in the manner described, presents the desirable physical features of the withdrawal of the heads of the alleged pin from the path of the j’arn fiber, without weakening the walls of the slot, and the tightening of the pin in the holes in the side walls without heading or spreading the ends of the pin at the bottom of the depression, thus avoiding the bur or roughness that often attaches to such a heading.

Appellant contends that the needle thus produced presents the identifying physical characteristics of a depression in the sides of the needle adjacent to the pin hole, longitudinal compression or shortening of the pin, which serves to tighten the same in its bearings in the sides of the pin, and compression of the metal of the said sides adjacent to said pin, by which the pin is further tightened in its bearings and the metal around said pin made denser and stronger, instead of being weakened by the removal of metal to make countersinks. This, he says, is the claim for an article or product of invention identified without the aid of the method by which it was produced. Without pausing to consider what force there may be in the criticism that, after all, the pin is made *720that way; that is, the way described in the specifications, and therefore can only be thus identified, we proceed to consider the alleged infringing needle of the defendant.

After a careful examination of the record, we accept as a correct description of the defendant’s needle, that given by defendant’s expert, at page 437 of the record, in connection with his exhibit drawing, at page 604:

“The needle blank before slotting has countersinks made in its sides as shown in B B' in Fig. 1 of the drawing which I now submit. After these countersinks aré constructed, the body A of the needle body has the slot A made in the middle thereof, leaving two sides A3 and A3, the hole D is then pierced through the side bars of the needle A3 A3, these two stages of construction being shown in. Figs. 2 and 3. Fig. 4 shows the latch F and pivot pin E inserted in the needle. Fig. 5 shows the ends of the pivot pin E riveted down to fill the' countersinks B with some surplus material Braised above the surface of the side bars of the needle. Fig. 6 shows this surplus material E2 removed from the head of the pivot pin, which is now flush with the side bars of the needle. Fig. 7 shows the pivot pin compressed; the imprints of the compressing die are shown at H, the heads of the pivot pin E2 being driven into the body of the side bars of the needle. Below the head of the pivot pin E2 is shown a portion of the pivot pin of its original diameter, while the expanded center of the pivot pin filling the hole in the latch of the needle is shown at Es. I have noticed in my examination of these pivot pins that the expanded central portion thereof shown at E2 usually begins somewhat within the body of the side bars of the needle, and that the inner ends of the holes in the side bars A3 are usually somewhat expanded around their orifice by the swelling of the pin, and some of the metal of the side bars is carried a little above the inner surface of the side bars, as shown at G. The effect of the compression upon the heads is usually to transform the countersink from a straight-sided cone, as shown at B, Fig. 6, to a slightly convex or bulging shape, as shown at E3, Fig. 7. The side view of the pivot of the finished needle is shown at Fig. 8.”

That there are certain differences between the physical features of the needle thus described and that of the patent in suit, we think must be apparent from the reading of -this description. These differentiations are, chiefly, large countersinks (whether made by compression or cutting away is not important), and the filling of these countersinks by the upsetting or heading of the ends of the pin. We think also it must be apparent that the heading or riveting of the pins, by which the countersinks are almost filled, was done prior to the compression applied to the heads of the pin and the surrounding metal of the sides of the needle and countersinks. Whether it is apparent, as a physical feature of the needle, that, after the riveting of the head there was this compression longitudinally of the pin and transversely of a portion of the metal in the sides of the needle, we do not know. We do know that the ends of the pin are held with measurable security in the walls of the socket, and that the central portion of the pin in the socket is expanded. To what extent this tightening of the ends of the pin in the hole and expansion of its central portion was due to the longitudinal compression of the pin and of the metal circumjacent to the holes, or to what extent it was due to the previous heading or riveting of the pin, we do not know. There was evidence which we think was sufficient to show that the heading or riveting of the pin in the counter*722sink would of itself serve to tighten and secure the ends of the pin in the side walls of the slot.

*721

*722If the three claims of the patent in suit are to be held to cover needles with pins that are headed, as well as those which are headless, then, aside from the prior art and from other evidence in the case, the defendant’s needle is covered by these claims. Though much may be said for the argument made by the learned counsel for the appellee, that it is inevitably deducible from the patent in suit, that the patented structure is limited to the presence therein of a headless or substantially headless pivot pin, we prefer to deal with those limitations which we think the proceedings in the Patent Office and the prior art placed upon the claims of the patent.

The letters patent in suit note an application filed August 26, 1901. The last paragraph but one of the specification states that “this application is a continuation of and has been filed to take the place of my former application, filed November 8, 1899,” and it appears, by reference to the file wrapper in the Patent Office, that this was allowed. If this is to be so considered, the proceedings in the Patent Office, under the original application of November, 1899, and the statements of the applicant in regard to the subject-matter of his claimed invention, become technically pertinent in the present case. We say “technically pertinent,” because, whether the application under which the patent in suit was granted be called a formal continuation, or not, of the original application for a patent with reference to the sanje subject-matter, the admissions and statements of the applicant in the former case are not irrelevant to the issues now under consideration, and we are not to be asked to close our eyes to evidential facts upon a ground so slight as the one suggested.

' The claims filed with the original application were fottr in number, three relating to the needle as a structure and one for the process of making it. The Patent Office refused to allow the process claim to be joined with the product claims, and the process claim No. 4 was therefore withdrawn. There is substantial similarity between the drawings submitted with this original application and the first four drawings of the patent in suit. The product claims were afterwards rejected, on the German patent to Mittlacher. This rejection brought about a correspondence between the Patent Office and applicant’s counsel, and in a letter dated October 1, 1900, the latter, after proposing various amendments, among other things states, with regard to the Mittlacher patent:

“In this construction it will be apparent that when the rivet is pressed endways to shorten the same, that the ends of the rivets will spread out over the' bottom of the countersink, so that the result will not be a needle with the ends of the rivet ‘flush with the bottoms of the countersinks,’ as claimed in claim 1, or the extremities of the rivet ‘coinciding with the bottoms of the countersinks,’ as specified in claim 2, or ‘the bottoms of the countersinks and the ends of the rivets occupying a position in substantially the same plane,’ as specified in claims 3 and 4. Instead, the rivet will be headed over upon the bottoms of the countersinks, so that the ends of the rivets and the bottoms of the countersinks will be in different planes which are the distance apart equal to the thickness of the head of the rivet.’’

*723And later, with reference to the pivot pin and the sides of the needle circumjacent to the hole, he says :

"Tlie metal of one does not crawl over or overlap the metal of the other, as is tlie case in both the methods shown in tlie German patent.”

Again, in another letter to the Patent Office, dated January 26, 1901, and cited by counsel for appellee, the applicant says:

“If the honorable examiner will kindly examine the last paragraph of page 4 and Tlie latter part of page <3, together with Figure 7 of the drawings, he will see that in the knitting machine needle in this application, there are no heads in the ends of the pivot pin. That is, that portion of the ends of the pivot pin which is confined in the side walls of the needle body, is of substantially uniform size, and the end of the pivot pin forms tlie central portion of tlie bottom of the countersink.”

Further on, he says:

"The pivot pin, therefore, of Hulee differs from the pivot pin of Mr. Currier's application, in that its ends, which are confined in the side walls of: the needle body, are not of substantially uniform size, but are instead, formed with heads or enlarged portions which fill the countersinks as in Figure 2. The specification of Mr. Currier especially states that the ends of tlie pivot pins are not upset in such a way as to make heads to stand in the countersinks.”

These careful differentiations of what the applicant claimed as his invention from the needles of the prior art, applies with great exactness to the needle described and claimed in the patent in suit, exclusive, perhaps, of the second modification stated in the specifications. The correspondence in which these differentiations were made, resulted in the filing of four amended claims, the first three of which are identical, word for word, with the three claims of the patent in suit, except that the word “countersinks” is used, in place of the word “depressions,” as used in the claims of the patent in suit. All four of the claims were,, on June 1, 1901, declared objectionable by the examiner, on the ground mat the “needle as an article of manufacture does not appear to be distinguished, at least in a patentable sense, from that shown in the patent to Fluke et al. If there be anything patentable, it is in applicant’s method of forming his needle, which could not be claimed in this application.” What occurred between this action of the examiner and August 26, 1901, when a new application was filed, with specifications the same as appear in the patent now before us, and with five claims, the first three being the same as those now in the patent in suit and the fourth and fifth covering pivot pins having heads which fill the countersinks in the side walls of the needle, does not appear in the record.

It seems inexplicable, after the limitations above referred to, put by the applicant upon his amended claims in the proceedings connected with the original application and the rejection of those claims by the examiner, as above referred to, that he should, in this new application, in continuation of the old, put back those three claims, which he had so carefully limited and differentiated from needles having pins with headed or upset ends filling or partially filling the countersinks surrounding the holes, with the addition of two claims which covered the very needles from which he had so carefully differentiated Ms own invention. This cannot be explained on the theory that the first three *724claims are to be considered as generic, and the fourth and fifth for a species of the genus, because the applicant, in the correspondence referred to, has defined the invention covered by these three claims as needles with headless pins alone, and the headless feature, as much as any other, inheres in the generic character of the invention covered by the claims. But so it is, and upon this renewed application an inter-. ference was granted to one Beckert, of Chemnitz, Germany, the manufacturer of the defendant’s needles, and whose agent for selling the same the defendant is. Beckert’s application for a -patent' was filed August 4, 1900, prior to Currier’s renewed application, and his claims were’ substantially identical with its five claims. Currier, however, swore that he conceived the invention embodied in the first three claims —that is, of the headless pin construction — in 1897, but as to the invention expressed in claim 4 (which with claim 5 refers to small countersinks and conical countersinks, into which the heads of the pivot pins are upset or riveted, filling or partially filling the same), he says he did not conceive the same until January 26, 1901. Beckert could not carry his invention, as embodied in Currier’s first three claims, back of 1900, but he clearly anticipated the invention embodied in claims 4 and 5 of the Currier application. The adjudication of the Patent Office, therefore, in regard to the fourth (and fifth) claim was in Beckert’s favor, and in regard to the first three claims, was in favor of Currier. Afterwards, the fourth and fifth claims were withdrawn by Currier and the patent issued for the first three claims, which are the claims here under consideration.

In addition to the differentiations made in the correspondence above alluded to, we here have the applicant for the patent in suit, in his statement in the interference proceedings, plainly distinguishing, not merely specifically but generically, between the invention covered by the first three claims of his application, being the claims now of the patent in suit, and that covered by the fourth (and fifth), thus necessarily confining the three claims of the patent in suit to needles characterized by headless pivot pins. The very able counsel for the appellant seeks to avoid this conclusion, by the argument that, inasmuch as the three claims do not, in terms, refer to either headed or headless pivot pins, the invention of the patent in suit does not consist, nor is it dependent upon whether the opposite terminal ends of the pivot pins are headed, or whether they are headless; that either is a mere incident to the essential characteristic of the thing described and claimed; that either is simply a species of the genus, and it is the latter which constitutes the invention made and the invention claimed; and that the applicant’s disclaimer in the interference proceedings was merely with reference to the species, and not to the generic form exemplified by the headless structure.

The fallacy of this argument lies in a misconception of the facts upon which it rests. The differentiations made in the correspondence with the Patent Office above referred to, in regard to the stated references, is inconsistent with the distinction here attempted between “genus” and “species.” The elements which constitute the invention are certain characteristic physical features. It is a combination, in which *725each feature is an essential part. As said by the learned judge of the court below, “the use of compressing dies to force iu the heads and surrounding surfaces to form depressions, which is the only distinction asserted,” is in no respect new.

It is also to be noted, that the interference proceedings between Beckert and Currier left Beckert, on the ground of priority of invention, with full title to the pin described in claim 4 (and 5) of Currier’s last application. 'Phis result was not questioned, or in any way criticised by Currier. Beckert was left, therefore, with the unquestioned right, not only to use headless pivot pins, filling or partially filling the countersinks, in conjunction with some other form of needle than one having the pins longitudinally compressed simultaneously with a portion of the needle side circumjacent thereto, hut iu conjunction with that exact form as described in claim 4 of his own application, and in claims 4 and 5 proposed in the last application for the patent in suit, and afterwards withdrawn.

The history of the case, as above recited, makes the headless pins, compressed in the manner described, simultaneously with a portion of the circumjacent surface of the needle sides, the generic invention of the patent in suit, and the defendant’s needle, with its headed pins upset so as to fill or partially fill the countersinks circumjacent to the holes, is not covered by the claims thus limited and defined. Tittle need be-added to what has been said by the learned judge of the court below as to the effect of the public use proceedings upon the issue here involved. After the finding by the Patent Office, that Beckert was entitled to priority of invention, as to the needle structure covered by the fourth (and fifth) claim of Currier’s application, and this, upon the express statement by Currier that he had not conceived of the needle having headed pivot pins until January 26, 1901, public use proceedings were brought about by Currier and the appellants, as to needles with these same headed pins, and protesting against the granting of a patent to Beckert therefor. The protest concludes as follows:

“The ground of this jjrotest is, that knitting machine needles, containing the invention described in the fourth claim of Beckert’s application, have been in public use and on sale in the United States for more than two years prior to the filing of the above-entitled application of Beckert, and that this public use and sale has been notorious and extensive.”

The ground of this protest was sustained by the Commissioner, and the patent refused to Beckert. Appellants, however, can take nothing by this finding. Whatever view we take as to the question, whether or not the application of August, 1901, was a continuation of the original application of November, 1899, it seems dear that, after the withdrawal of claims 4 and 5 of the new application, the invention covered thereby was abandoned, and needles with headed pins, as described therein, became public property, and cannot tiow he monopolized by the appellants. But it is also clear that, though the application of August, 1901, may be a continuation of the original application of November, 1899, so far as the first three claims filed with this later application are concerned, it is an independent and new application, as regards the fourth and fifth claims thereof. The public use, therefore, carried back *726more than two j^ears prior to Beckert’s application of August 4, 1900, antedates the new application of Currier, of August, 1901, for a still longer period, and prevents, not only Beckert’s fourth claim, but the claims of the patent in suit, from covering needles with headed pins.

The decree of the court below is hereby affirmed.